Based on the linked source; automatically prepared and checked against the original report.
Mental tongue commands
Researchers introduced an EEG-based brain-computer interface that interprets imagined tongue movements rather than physical tongue actions, intraoral devices or a restricted set of offline commands. The approach is designed to provide intuitive, entirely mental control for assistive wheelchair technology, drawing on the tongue’s cortical representation and preserved motor function.
Signals and experimental design
The system used six imagined actions: touching the left or right mouth corners, touching the upper or lower lips, and creating left or right cheek bulges. These tasks were intended to produce alpha-band event-related desynchronization patterns over the tongue motor cortex. EEG recordings came from 15 healthy participants wearing a 14-channel consumer-grade EMOTIV EPOC X headset.
Accuracy and remaining work
Researchers extracted alpha-band features and tested linear discriminant analysis, support vector machine, naïve Bayes and artificial neural network classifiers. Two-class tasks reached 76.19% accuracy, while performance declined as task complexity increased; the artificial neural network performed best in multiclass scenarios. The abstract says improved classification, user training and real-time validation remain necessary for robustness and practical usability.
